A personalized mRNA cancer vaccine developed with the assistance of artificial intelligence has helped shrink a tumor in a dog diagnosed with advanced cancer. The experimental treatment was created through a collaboration between a technology entrepreneur and academic researchers, demonstrating how AI tools can assist in designing targeted medical therapies.
The effort began when Australian technology entrepreneur Paul Conyngham sought new treatment options after his dog Rosie was diagnosed with aggressive mast cell cancer in 2024. He decided to sequence DNA of the dog to find the problem and build a cure. And to do this, he relied on ChatGPT. He told The Australian, “I went to ChatGPT and came up with a plan on how to do this.” Paul Conyngham is no stranger to AI, having worked in the field of machine learning and data analysis for 17 years. Using AI systems including ChatGPT and AlphaFold, Conyngham analyzed genomic data from the dog’s tumor to identify mutations that could potentially be targeted by an experimental therapy.
The project began with DNA sequencing conducted at the University of New South Wales Ramaciotti Centre for Genomics. Researchers compared genetic material from the dog’s healthy tissue and tumor cells to identify mutations associated with the disease. Conyngham then used computational analysis pipelines to interpret the genomic data and locate specific protein targets linked to the cancer.
Using AlphaFold, a system developed to predict protein structures from genetic sequences, the analysis helped determine how mutations altered the structure of certain proteins. These insights were used to design an mRNA sequence intended to stimulate the dog’s immune system to recognize and attack cancer cells carrying those mutations.
The mRNA treatment was manufactured by researchers at the UNSW RNA Institute under the direction of nanomedicine specialist Pall Thordarson. The vaccine design was condensed into a short molecular sequence that was packaged into nanoparticles capable of delivering the genetic instructions into cells.
Messenger RNA vaccines work by instructing cells to produce specific proteins that trigger an immune response. During the COVID-19 pandemic, similar technology was used in vaccines developed to prevent viral infection. Researchers are now investigating whether the same platform can be adapted for cancer treatment by targeting tumor specific mutations.
After the vaccine was prepared, veterinary oncologist Rachel Allavena at the University of Queensland administered the treatment under an existing research ethics framework for experimental canine cancer therapies. Rosie received the first injection in December followed by a booster dose several weeks later.
According to researchers involved in the study, one of the dog’s large tumors decreased substantially in size after treatment. While the therapy has not eliminated the cancer entirely, veterinarians reported improvements in the animal’s comfort and activity levels.
Scientists say the case illustrates how advances in genomics and computational biology can accelerate the development of personalized treatments. By combining genetic sequencing, protein modeling, and AI assisted analysis, researchers were able to design a custom therapeutic candidate in a relatively short period.
The experiment also highlights the emerging concept of personalized oncology, in which therapies are designed around the specific genetic mutations present in an individual tumor. Several research groups are currently testing mRNA based cancer vaccines in human clinical trials.
Researchers caution that the canine case remains an experimental application and does not yet represent a validated medical treatment. However, the approach demonstrates how computational tools can assist scientists in identifying molecular targets and designing individualized therapies.
Further sequencing and analysis are underway to study why certain tumor regions did not respond to the initial treatment and to explore whether additional vaccine designs could improve results.
